Vasculitis in patients with inflammatory bowel diseases: A study of 32 patients and systematic review of the literature

Seminars in Arthritis and Rheumatism 45 (2016) 475–482

Contents lists available at ScienceDirect

Seminars in Arthritis and Rheumatism journal homepage: www.elsevier.com/locate/semarthrit

Vasculitis in patients with inflammatory bowel diseases: A study of 32 patients and systematic review of the literature Alice Sy, MDa, Nader Khalidi, MDb, Natasha Dehghan, MDc, Lillian Barra, MD, MPHd, Simon Carette, MDa, David Cuthbertson, MSe, Gary S. Hoffman, MDf, Curry L. Koening, MDg, Carol A. Langford, MDf, Carol McAlear, MAh, Larry Moreland, MDi, Paul A. Monach, MD, PhDj,k, Philip Seo, MDl, Ulrich Specks, MDm, Antoine Sreih, MDh, Steven R. Ytterberg, MDn, Gert Van Assche, MD, PhDo, Peter A. Merkel, MD, MPHh, Christian Pagnoux, MD, MSc, MPHa,n, Vasculitis Clinical Research Consortium (VCRC) and the Canadian Vasculitis Network (CanVasc) a Division of Rheumatology, Mount Sinai Hospital, University of Toronto, The Joseph and Wolf Lebovic Building, 60 Murray St, Ste 2-220, Toronto, Ontario, Canada M5T 3L9 b Division of Rheumatology, McMaster University, Hamilton, Ontario, Canada c Division of Rheumatology, University of British Columbia, Vancouver, British Columbia, Canada d Division of Rheumatology, St. Joseph’s Health Care, London, Ontario, Canada e Department of Biostatistics, University of South Florida, Tampa, FL f Center for Vasculitis Care and Research, Cleveland Clinic, Cleveland, OH g Division of Rheumatology, University of Utah, Salt Lake City, UT h Division of Rheumatology, University of Pennsylvania, Philadelphia, PA i Division of Rheumatology, University of Pittsburgh, Pittsburgh, PA j The Vasculitis Center, Section of Rheumatology, Boston University School of Medicine, Boston, MA k Clinical Epidemiology Unit, Boston University School of Medicine, Boston, MA l Division of Rheumatology, Johns Hopkins University, Baltimore, MD m Division of Pulmonary and Critical Care Medicine, Mayo Clinic College of Medicine, Rochester, MN n Division of Rheumatology, Mayo Clinic College of Medicine, Rochester, MN o Division of Gastroenterology, University of Toronto-Inflammatory Bowel Disease Clinic, Toronto, Ontario, Canada

a r t i c l e in fo

Keywords: Vasculitis Takayasu arteritis Inflammatory bowel disease Crohn’s disease Ulcerative colitis

a b s t r a c t Background: Published small case series suggest that inflammatory bowel disease [IBD; Crohn’s disease (CD) or ulcerative colitis (UC)] and vasculitis co-occur more frequently than would be expected by chance. Objectives: To describe this association by an analysis of a large cohort of carefully studied patients and through a systematic literature review. Methods: Patients with both IBD and vasculitis enrolled in the Vasculitis Clinical Research Consortium (VCRC) Longitudinal Studies, followed in Canadian Vasculitis research network (CanVasc) centers and/or in the University of Toronto’s IBD clinic were included in this case series. A systematic literature review of patients with IBD and vasculitis involved a PubMed search through February 2014. The main characteristics of patients with Takayasu arteritis (TAK) and IBD were compared to those in patients with TAK without IBD followed in the VCRC. Results: The study identified 32 patients with IBD and vasculitis: 13 with large-vessel vasculitis [LVV; 12 with TAK, 1 with giant cell arteritis (GCA); 8 with CD, 5 with UC]; 8 with ANCA-associated vasculitis [AAV; 6 granulomatosis with polyangiitis (GPA), 2 with eosinophilic granulomatosis with polyangiitis (EGPA)]; 5 with isolated cutaneous vasculitis; and 6 with other vasculitides. Patients with LVV and AAV were mostly female (18/21). The diagnosis of IBD preceded that of vasculitis in 12/13 patients with LVV and 8/8 patients with AAV. The review of the literature identified 306 patients with IBD and vasculitis: 144 with LVV (133 TAK; 87 with IBD preceding LVV), 19 with AAV [14 GPA, 1 EGPA, 4 microscopic polyangiitis (MPA)], 66 with isolated cutaneous vasculitis, and 77 with other vasculitides. Patients with IBD and TAK

Financial support: Alice Sy was funded by the Abbott Summer studentship granted by the Canadian Rheumatology Association. The Vasculitis Clinical Research Consortium has received support from the National Institute of Arthritis and Musculoskeletal and Skin Diseases, USA (U54AR057319 and U01AR5187404), the National Center for Research Resources, USA (U54RR019497), the Office of Rare Diseases Research, and the National Center for Advancing Translational Science, USA. n Corresponding author. E-mail addresses: [email protected], [email protected] (C. Pagnoux). http://dx.doi.org/10.1016/j.semarthrit.2015.07.006 0049-0172/& 2015 Elsevier Inc. All rights reserved.

476

A. Sy et al. / Seminars in Arthritis and Rheumatism 45 (2016) 475–482

were younger and had more frequent headaches, constitutional symptoms, or gastrointestinal symptoms compared to those patients in the VCRC who had TAK without IBD. Conclusions: These findings highlight the risk of vasculitis, especially TAK, in patients with IBD (both CD and UC). & 2015 Elsevier Inc. All rights reserved.

Vasculitis comprises a group of rare and potentially lifethreatening diseases typically classified by the size of the vessels predominantly affected [1]. Primary large-vessel vasculitides (LVV) affect the aorta and its branches and include Takayasu arteritis (TAK) and giant cell arteritis (GCA). Medium vessel vasculitides affect the main visceral arteries and their branches and include polyarteritis nodosa (PAN) and Kawasaki disease. Small-vessel vasculitides affect arterioles and/or capillaries and include antineutrophil cytoplasmic antibody (ANCA)-associated vasculitides [AAV, including granulomatosis with polyangiitis (GPA), eosinophilic granulomatosis with polyangiitis (EGPA) and microscopic polyangiitis (MPA)] and immune-complex-mediated vasculitides [1]. Gastrointestinal tract manifestations can occur in these vasculitides [2,3] due to involvement of the mesenteric arteries or smaller vessels, potentially leading to bowel ischemia. Granulomatous inflammation of the bowel mucosa can develop in GPA and EGPA and occasionally mimics inflammatory bowel diseases [IBD; Crohn’s disease (CD) or ulcerative colitis (UC)] [3–5]. The literature on vasculitis in patients with IBD is quite limited. The current understanding of this rare association stems predominantly from case reports and small case series with not more than 10 patients. Various vasculitides have been associated with IBD, especially TAK [6–8] but also GPA [9], isolated cutaneous vasculitis [10], and central nervous system vasculitis [11]. The first objective of this study was to describe the different types of vasculitis that can occur in association with IBD, the temporal relationships of these diagnoses, and the management and clinical outcomes of patients with these 2 diseases. The second objective was to conduct a comprehensive review of the literature on the association of vasculitis and IBD. Additionally, the clinical characteristics of patients with IBD and TAK were compared to those with TAK without IBD.

2 other sources, the participating physicians identified patients followed in their clinics who met the inclusion criteria. Because it was expected that TAK would be the main type of vasculitis associated with IBD, the characteristics of patients with TAK and IBD were compared to those of patients with TAK without IBD from the VCRC cohort. The protocol was reviewed and approved by the VCRC Steering Committee and by the local research ethics boards at all participating centers. Literature search and study selection Two authors (A.S. and C.P.) conducted a systemic literature review to identify published cases of patients with IBD and vasculitis. MEDLINE was searched via PubMed for reports published from 1964 to February 2014, without a language restriction, with the following keywords: Takayasu arteritis, aortitis, giant cell arteritis, large-vessel vasculitis, anti-neutrophil cytoplasmic antibodyassociated vasculitis, Wegener’s granulomatosis, Churg–Strauss syndrome, small-vessel vasculitis, retinal vasculitis, systemic vasculitis, central nervous system vasculitis, cerebral vasculitis, cutaneous vasculitis, or angiitis in combination with inflammatory bowel disease, ulcerative colitis, or Crohn’s disease. Retrieved and relevant papers were manually searched for additional references. Studied parameters

Material and methods

The following information, when available, was collected for each patient, directly from patient charts or from retrieved articles, by using a standardized data collection form: age at diagnosis, sex, comorbidities, timing of the diagnoses of IBD and vasculitis, main clinical manifestations of vasculitis, routine laboratory results, medications, and follow-up outcomes (survival and disease status). Physicians within the VCRC, CanVasc, and UoT-IBD were contacted directly to provide information on any missing data.

Patients

Statistical analysis

Patients were included in the study if they had vasculitis and IBD. IBD could have developed either before or after the former. Only patients with Behçet’s disease or with AAV diagnosed within the same year as IBD were excluded, because inflammatory colitis due to these vasculitides can be misdiagnosed as IBD [12,13]. Patients were identified from 3 different sources: the Vasculitis Clinical Research Consortium (VCRC) Longitudinal Studies cohorts, the Canadian Vasculitis research network (CanVasc) centers, and the Mount Sinai Hospital–University of Toronto referral center for IBD (UoT-IBD). The VCRC Longitudinal Studies prospectively collect data in patients with TAK, GCA, PAN, EGPA, GPA, or microscopic polyangiitis (MPA), all satisfying the 1990 American College of Rheumatology (ACR) modified criteria [14,15], from 7 referral centers in the United States and 2 in Canada (http://www.raredi seasesnetwork.org/VCRC). The CanVasc is a Canadian network for research on vasculitis involving clinics in 18 cities (http://www. canvasc.ca), including the 2 centers also in the VCRC. In the VCRC database, IBD is systematically recorded as a comorbid illness when present, either at baseline or during follow-up visits. For the

Categorical variables are expressed with number (percentage) and comparisons using the chi-square test or, when appropriate, Fisher’s exact test. For continuous variables, median (range) are provided, and means were compared by Student’s t test. A p o 0.05 was considered significant.

Results Patients from the VCRC, CanVasc, and UoT-IBD Overall, 43 patients (21 from the VCRC and 22 from CanVasc, including 2 co-managed with the UoT-IBD) with IBD and vasculitis diagnosed between 1986 and 2013 were identified. In all, 4 patients from the VCRC (2 EGPA and 2 GPA) and 5 from CanVasc (4 GPA and 1 EGPA) were excluded because IBD and vasculitis were diagnosed within 1 year of each other. A patient with GCA and another with EGPA were excluded because a definitive diagnosis for colitis could not be established. The demographics and clinical

A. Sy et al. / Seminars in Arthritis and Rheumatism 45 (2016) 475–482

characteristics of the final 32 patients (17 from the VCRC and 15 from CanVasc) are summarized in Table 1. Patient 5 from Table 1 had been reported in a separate case report (also included in the literature review) but follow-up is now longer [16]. The most common group of vasculitides with IBD identified was LVV (n ¼ 13), consisting of 12 cases of TAK and 1 case of GCA (8 with CD and 5 with UC). Other cases included 8 cases of ANCAassociated vasculitis (6 GPA with anti-proteinase 3 ANCA and 2 EGPA), 5 with isolated cutaneous vasculitis, and 6 with other vasculitides. Most patients with LVV and ANCA-associated vasculitis were female [median age 27 (range: 17–58) years and 20 (range: 8–52] years at the time of diagnosis of vasculitis and IBD, respectively). The diagnosis of IBD preceded that of LVV for 12/13 patients and for all 8 patients with ANCA-associated vasculitis. The median intervals between the diagnoses of vasculitis and IBD was 10 years (2 months to 27 years) for LVV and 11 years (2–44 years) for ANCA-associated vasculitis. The most common clinical findings in patients with TAK and IBD were constitutional symptoms (in 12/12 patients), vascular signs or symptoms (in 10/12 patients), including limb claudication, decreased peripheral pulse, blood pressure differences between arms, and bruits. Detailed results of computerized tomographic or magnetic resonance arteriography findings were available for 9/12 patients: 5 showed an aneurysm or dilation of at least a portion of the aorta; 7 had stenosis involving the branches of the thoracic aortic artery, 6 showed renal/suprarenal artery involvement, 4 had celiac, superior or inferior mesenteric artery involvement, and 2 had vertebral artery involvement. The following medications were used in isolation or combination to treat the majority of the patients with IBD: 5-aminosalicylates (5ASA), glucocorticoids, methotrexate, azathioprine, or an anti-tumor necrosis factor (TNF) alpha agent; 12/32 patients (38%) received at least 1 biologic, most commonly infliximab. A total of 4 patients had also undergone large bowel surgery; 2 patients had previously received infliximab. In 5 patients with CD, TAK developed despite infliximab, but in none of the other 11 patients did the vasculitis develop while receiving a biologic for their IBD. Of the 29 patients with information on their initial treatment regimen for the vasculitis, 27 (93%) received intravenous or oral glucocorticoids, 23 (79%) received another immunosuppressant such as azathioprine, methotrexate or cyclophosphamide, and 1 (3%) underwent plasma exchange. Overall, 6 (21%) patients received biologics, including infliximab, adalimumab, or abatacept. Follow-up information was available for 28/32 patients (88%). In all, 3 patients were newly diagnosed with vasculitis and 1 from the VCRC had no documented follow-up visit after the diagnosis of vasculitis at the time of this analysis. After a median follow-up of 5.4 years (range: 3 months–28 years) after the diagnosis of vasculitis, 17 patients (61%) had achieved remission of the vasculitis and 11 (39%) had persistent disease activity or experienced vasculitis flare(s). In 5/9 patients (56%) who received biologic therapy for either vasculitis or IBD and for which follow-up data were available, the vasculitis was in remission at the last follow-up appointment. Literature review From the literature review (Fig. and Appendix 1), 306 patients with IBD and vasculitis were identified, within the same 4 main disease categories (LVV, ANCA-associated vasculitides, cutaneous vasculitis, and other vasculitides; Table 2). Patients with IBD and LVV The 144 cases of LVV included 70 cases of CD and 74 cases of UC. Most patients (82%) were female; 133 (92%) were reported as

477

having TAK and 11 (8%) patients as having non-TAK LVV (GCA or other LVV). Of the 127 cases with data on the temporal relationship between the 2 diagnoses, the diagnosis of IBD preceded that of vasculitis in 87 (69%) by a median of 4 years (range: 0.5–31 years), including 1 case of LVV with UC and 3 with CD that were reported as possibly drug-induced. In 22 patients, LVV preceded IBD by a median of 6 years (range: 3 months–36 years); vasculitis and IBD were diagnosed concurrently in the remaining 18 patients. Fewer than 10% of the reported cases with CD and TAK but 66% of those with LVV and UC were from Japan. There were no other significant differences between patients with LVV with UC or CD. Most patients received a combination of 5-ASA, glucocorticoids, and other immunosuppressive drugs. Overall, 20 patients (16 with CD and 4 with UC) received biologics (16 infliximab, 2 infliximab then adalimumab, 1 adalimumab, and 1 tocilizumab). Good clinical response with biologics was noted for 11/20 patients (55%), which allowed for tapering or discontinuation of the glucocorticoids. In all, 2 of the remaining 9 patients experienced severe allergic reactions to infliximab and 1 patient each had persistently active CD and required hemicolectomy, experienced recurrent infections, or developed aortic arch aneurysm despite infliximab treatment. In the 4 remaining patients, TAK developed while receiving antitumor necrosis factor (TNF)-alpha agents, which were later stopped, and all patients showed improvement after receiving high-dose glucocorticoids alone. In all, 9 of the 70 patients (13%) with CD and TAK and 7 of the 73 patients (10%) with TAK and UC required a bowel resection for their IBD. The median follow-up of patients with CD and LVV was 0.7 years (range: 1 week–20 years) and 0.2 years (range: 1 week–13 years) for patients with UC and LVV. Among these patients, 2 with CD died (myocardial infarction and non-specified cause), as did 4 with UC (one each from cerebral vascular event, aneurysm rupture, respiratory failure, and non-specified cause).

Patients with IBD and other types of vasculitis In total, 66 patients had IBD and cutaneous vasculitis: 45 had been reported with isolated cutaneous vasculitis (25 with “leukocytoclastic vasculitis” and 20 with cutaneous PAN), 11 with isolated cutaneous granulomatous small-sized vessel vasculitis, 8 with IgA vasculitis (Henoch–Schönlein purpura), and 1 each with nodular vasculitis, urticarial vasculitis, and lymphocytic vasculitis. For 33/ 56 cases (59%) with cutaneous vasculitis and available information on the intervals between the 2 diagnoses, the vasculitides occurred after the diagnosis of IBD. All but 3 (95%) of these cutaneous vasculitides reportedly resolved after treatment with glucocorticoids, mesalamine, sulfasalazine, azathioprine, methotrexate, or dapsone, but some patients were additionally treated with antibiotics or colectomy when the IBD was active. Overall, 8 cases (including 3 with IgA vasculitis) were reported as secondary to anti-TNF-alpha treatment (infliximab or adalimumab). Cutaneous lesions resolved after treatment with methylprednisolone, diphenhydramine, and ranitidine in 1 patient and, for 6/7 of the remaining patients, after the discontinuation of the anti-TNF-alpha therapy; however, the re-administration of the same anti-TNFalpha therapy in 2 patients was associated with recurrence of skin lesions. In all, 19 patients with ANCA-associated vasculitides were reported: 14 with GPA, 1 with EGPA, and 4 with MPA. Of the 16 patients with available information on treatments and outcomes, 9 had received glucocorticoids combined with cyclophosphamide (followed by mycophenolate mofetil, azathioprine, or methotrexate), azathioprine, methotrexate, or infliximab; 11 reportedly entered remission, and “good response” to treatment was obtained in the remaining 5. A total of 77 other cases of vasculitis were found associated with IBD, including more than 10 cases each of IgA vasculitis with

478 Table 1 Main demographic data and clinical characteristics of 32 patients with both IBD and vasculitis enrolled in the Vasculitis Clinical Research Consortium (VCRC) longitudinal studies, followed in Canadian Vasculitis research network (CanVasc) centers and/or in the University of Toronto’s IBD clinic Patient ID

Sex Inflammatory bowel disease Type Age at dx (years)

Large-vessel vasculitis 1 F CD 24 2 F CD 52 3 F CD 25 M F F

CD CD CD

13 15 8

7

F

CD

9

8 9 10 11 12 13

F F F F F M

CD UC UC UC UC UC

27 16 20 18 34 30

ANCA-associated vasculitis 14 F CD 8

Treatment

Type

Age at dx (years)

Clinical characteristics

Treatment

Length of follow- Disease status and treatment up (years) at last follow-up

5-ASA AZA, hemicolectomy PDN, AZA, IFX, ADA

TAK TAK TAK

27 58 25

CONST, VASC, MSK, ART (þ ) CONST, VASC, MSK, ART (þ ) CONST, ARTH, ART (þ )

PDN, AZA PDN, AZA PDN, AZA, IFX

28 7.5 5.5

IFX 5-ASA, PDN, IFX BUD, IFX, ileostomy, subtotal colectomy AZA, IFX, ADA subtotal colectomy 5-ASA, PDN, AZA, IFX 5-ASA, PDN, 6-MP 5-ASA 5-ASA, PDN, 6-MP PDN PDN

TAK TAK TAK

17 17 17

CONST, HTN, VASC, MSK, ART ( þ ) CONST, VASC, ART (þ ) CONST, HTN, VASC, PNS, SEM, CoA, ART (þ )

1 NA 4.5

TAK

20

CONST, VASC, MSK, ART (þ )

PDN, MTX, IFX MTX PDN, AZA, MTX, RTX IFX, LEF, ADA, ABT, aortic stent PDN, MTX

(new dx)

Recent flare; on PDN, MTX In remission; on AZA In remission; on PDN, AZA, IFX In remission; on PDN In remission; on PDN Recurrent flares; on PDN, MTX, IFX on PDN, MTX

TAK TAK TAK TAK TAK GCA

27 29 42 38 49 57

CONST, CONST, CONST, CONST, CONST, CONST,

PDN, AZA, MTX, IFX PDN, ADA PDN, MTX – PDN PDN

8 1 4 (new dx) (new dx) 1.5

NA Active disease; on PDN, ADA In remission; on PDN, MTX – on PDN In remission; on PDN

GPA

20

PDN, MTX, Plasma exchange

(new dx)

PDN, MTX, Plasma exchange

GPA

39

CONST, MSK, CUT, ocular, ENT, PUL, atypicalANCA (þ ), PR3 (þ ), biopsy (þ ) CONST, MSK, CUT, renal, ENT, c-ANCA (þ ), PR3 (þ ) CONST, CUT, PUL, c-ANCA (þ ), PR3 (þ ) CONST, MSK, ocular, ENT, c-ANCA (þ ), PR3 (þ ) CONST, ENT, PNS, PUL, C-ANCA (þ ), PR3 (þ ), biopsy (þ ) CONST, MSK, renal, ocular, ENT, C-ANCA (þ ), PR3 (þ ), biopsy (þ ) CONST, MSK, CUT, PNS, asthma, PUL, c-ANCA (þ ) CONST, ocular, ENT, asthma, PUL, cardiac, p-ANCA (þ )

15

F

UC

33

5-ASA, 6-MP, MTX, IFX, ADA 5-ASA

16 17 18

M F F

UC UC UC

58 5 40

5-ASA NA 5-ASA, PDN, colectomy

GPA GPA GPA

71 55 50

19

F

UC

52

5-ASA

GPA

54

20 21

F F

CD UC

29 14

5-ASA 5-ASA

EGPA EGPA

65 17

Cutaneous vasculitis 22 M CD 26

PDN, AZA, IFX

LCV

41

23

F

CD

43

MTX

LCV

24

M

CD

38

PDN, IFX

25 26

M F

CD UC

59 30

Other vasculitis 27 M CD 28

M

29

M

HTN, VASC, renal, ART (þ ) VASC, ART (þ ) VASC, CNS, ART (þ ) ART (þ ) VASC, PNS, ocular, ART (þ ) PNS, PMR, ocular, ENT, biopsy ( þ )

PDN, MTX, AZA, CYC PDN, MTX PDN, MTX PDN, CYC

9

Low disease activity; on PDN

1 2 14

In remission; on PDN Recent flare; on PDN, RTX In remission; on PDN

PDN, AZA, CYC

1

In remission; on PDN, AZA

PDN, AZA, CYC PDN, AZA, MTX

6 10

In remission; on PDN, MMF In remission; off treatment

CUT, biopsy (þ )

PDN, IFX

0.5

51

CONST, CUT, biopsy (þ )

PDN, MTX

LCV

49

CONST, MSK, CUT, biopsy ( þ )

PDN

0.5

5-ASA 5-ASA

LCV Cutaneous PAN

59 21

CONST, CUT, biopsy (þ ) CUT

PDN, MTX PDN, MTX, AZA, MMF

4 11

Active disease; on AZA, IFX, starting COL Active disease; on PDN, MTX, will begin IFX Active disease; on PDN, will begin IFX In remission; on MTX In remission; on MTX

13

5-ASA, PDN, MTX, IFX

Kawasaki

10

NA

NA

6

CD

13

PDN

34

Renal, biopsy (þ )

COL

6

CD

47

MTX, IFX

IgA vasculitis PAN

42

CONST, MSK, PNS

PDN, MTX

1

6

In remission; on IFX and MTX for CD Recurrent uveitis; considering MTX In remission; on MTX, IFX

A. Sy et al. / Seminars in Arthritis and Rheumatism 45 (2016) 475–482

4 5 6

Vasculitis

5-ASA, PDN, AZA, IFX 25 UC F 32

5-ASA, 5-aminosalicyclic acid; 6-MP, mercaptopurine; ABT, abatacept; ADA, adalimumab; ANCA, anti-neutrophil cytoplasmic antibody; ART, arteriography (þ , positive with arterial abnormalities); AZA, azathioprine; BUD, budesonide; CD, Crohn’s disease; CoA, coarctation of aorta; COL, colchicine; CNS, central nervous system; CONST, constitutional symptoms; CUT, cutaneous; CYC, cyclophosphamide; dx, diagnosis; EGPA, eosinophilic granulomatosis with polyangiitis; ENT, ear nose throat; GCA, giant cell arteritis; GPA, granulomatosis with polyangiitis; HCQ, hydroxychloroquine; IFX, infliximab; IgA, immunoglobulin A; LCV, leukocytoclastic vasculitis; LEF, leflunomide; MMF, mycophenolate mofetil; MSK, musculoskeletal symptoms; MTX, methotrexate; NA, not available; PAN, polyarteritis nodosa; PDN, prednisone; PMR, polymyalgia rheumatica; PNS, peripheral nervous system; PUL, pulmonary; RTX, rituximab; SEM, systolic ejection murmur; TAK, Takayasu arteritis; UC, ulcerative colitis; VASC, vascular.

Multiple strokes; on clopidogrel 4 None 30

MSK, CUT, VASC, CNS

6 PDN, MTX F 31

CD

18

PDN

CNS vasculitis CNS vasculitis

49

VASC, CNS, ocular

1 F 30

UC

48

5-ASA, PDN, MMF

PAN

47

CONST, VASC, CNS, PNS, cardiac

PDN, MTX CYC, HCQ, COL

Recurrent strokes; on PDN, MMF Stable disease; on MTX, PDN

A. Sy et al. / Seminars in Arthritis and Rheumatism 45 (2016) 475–482

479

glomerulonephritis, retinal vasculitis, or central nervous system (CNS) vasculitis. Sporadic cases of systemic PAN, anti-glomerular basement membrane (anti-GBM) disease, Kawasaki disease, or isolated vasculitic neuropathy were described associated with IBD. Characteristics of patients with TAK with or without IBD Table 3 summarizes the demographic and clinical features of the 12 patients with IBD and TAK in this series (8 from the VCRC cohort), the 132 patients identified in the literature, and the 152 patients from the VCRC with TAK but without IBD. Most patients with IBD and TAK from the literature were Asian and had a much shorter follow-up than those from this series or from the VCRC but without IBD. Patients with IBD and TAK were younger at the time of diagnosis of TAK than those without IBD and had more frequent headaches, constitutional, and gastrointestinal symptoms. At the last follow-up, most patients (63–89%), achieved remission of their vasculitis, with similar rates in all groups. No patients with TAK from the VCRC cohort, with or without IBD, had died at the time of this analysis, as compared with 6 (9%) of those with TAK and IBD in the literature.

Discussion This case series, one of the largest reported to date, and the systematic literature review emphasize that different types of vasculitis can occur in patients with IBD, including in decreasing order of frequency: LVV, mainly TAK, cutaneous vasculitis, and ANCA-associated vasculitides, mostly GPA. In most cases, the diagnosis of IBD preceded that of vasculitis by years, and IBD was usually not active at the time of the onset of vasculitis. TAK and ANCA-associated vasculitis are rare and the probability of their coexistence with IBD by chance is minimal, although this occurrence cannot be excluded. The prevalence of CD was as high as 9% in a French series of 44 patients with TAK [6]. In a North American study of 32 patients with TAK, 2 (6%) had CD, as compared to an estimated prevalence of 0.2% of CD in the general population [17]. Of the 160 patients with TAK in the VCRC, 8 (5%) patients with associated IBD were identified. The reasons for the more distinctive association between IBD and TAK or GPA are unclear but may indirectly support some common pathogenic mechanisms [18]. TAK and UC were also found to share a significant proportion of their genetic background, which may contribute to their coexistence [19,20]. A recent genetics study showed that patients with TAK complicated with UC displayed higher positivity for HLA-Bn52:01 than patients without UC [20]. Granulomatous inflammation is a hallmark of both TAK and GPA, as is also a key feature of CD [21,22]. There were some demographic and clinical characteristics more common among patients with coexisting vasculitis and IBD. Patients with IBD and LVV or ANCA-associated vasculitis were often female. Most patients with IBD and TAK in the literature were Asians but not in the current series, which included patients from North American vasculitis networks. Rény et al. [6] reported that patients with coexistent CD and TAK tended to be younger at the time of diagnosis of TAK and had more frequent constitutional symptoms, including fever, weight loss, or fatigue, than those with TAK without IBD. The current study replicated these findings. Patients with IBD and TAK from the literature and this series were on average 8–12 years younger at the time of the diagnosis of TAK than those with TAK without IBD. Patients with IBD and TAK were also more likely to have constitutional symptoms, gastrointestinal symptoms, and headache/dizziness but were less likely to experience claudication of the extremities. A recent case series of 11 patients with ANCA-associated vasculitides (7 patients with GPA

480

A. Sy et al. / Seminars in Arthritis and Rheumatism 45 (2016) 475–482

Articles identiied from Pubmed and by manual search through the reference lists x = 1054 Articles documenting at least one case of IBD associated with vasculitis x = 248

Cases of vasculitis associated with IBD n = 306*

Cases of large vessel vasculitis associated with IBD

Cases of cutaneous vasculitis associated with IBD

x = 114 n = 144

Cases of ANCAassociated vasculitis associated with IBD

x = 56 n = 66

Other vasculitides associated with IBD

x = 12 n = 19

x = 66 n = 77

Fig. Algorithm of the literature search (ANCA, anti-neutrophil cytoplasmic antibody; n, number of cases; x, number of articles).

and CD and 4 with EGPA and UC) suggested that the coexistence of GPA and UC or EGPA and CD was infrequent [22]. We did not observe this exclusive associative pattern in our patients (1 GPA with CD, 3 GPA with UC, 1 EGPA with CD, and 1 EGPA with UC), but our study replicates another important finding—that the diagnosis of IBD most often antedates the diagnosis of ANCA-associated vasculitis. Cutaneous vasculitis in patients with IBD remains an isolated and less severe complication, which is sometimes seemingly caused by the drugs used to treat IBD. Although less common, IgA vasculitis with or without glomerulonephritis, retinal vasculitis, and central nervous system vasculitis have also been associated with IBD. Some of these diseases could be coincidental and/or drug-induced. Importantly, patients with IBD carry an increased risk of thrombotic events, including strokes and white-matter CNS abnormalities [23,24], which may be mistaken as CNS vasculitis. Therefore, reported cases of CNS vasculitis, including those seen in this series, may be questionable without histological evidence of vasculitis.

Interestingly, the same drugs are often used to treat IBD and vasculitis, especially TAK. Anti-TNF-alpha monoclonal antibodies have been found effective for cases of IBD-associated TAK that did not adequately respond to conventional medical treatment and immunosuppressants [18]. In the current study, vasculitis developed in only 1/12 patients receiving anti-TNF-alpha therapies; however, several cases of new-onset TAK or cutaneous vasculitis have been reported in patients with IBD receiving these drugs [16,25–29]. This study has several strengths, including the relatively large sample size of 32 patients with vasculitis and IBD with substantial time of follow-up, a systematic literature review, and the inclusion of all types of vasculitides. Most previous studies focused on only 1 type of vasculitis [9]. Importantly, patients with a diagnosis of Behçet’s disease or a diagnosis of ANCA-associated vasculitis made within the same year were excluded from this study because they represent a different population, with possible initial misdiagnoses. The initial presenting features of IBD and these vasculitides can overlap and delay or confound the diagnosis of one or the

Table 2 Summary of the cases of vasculitis associated with IBD in this series and reported in the literature Large-vessel vasculitides This series No. of cases No. of Crohn’s disease (%) No. of ulcerative colitis (%) Sex, female/male (%) Median age at diagnosis of IBD, years (range) Median age at diagnosis of vasculitis, years (range) Temporal relationship IBD firsta Vasculitis first Concurrent

13 8 (62%) 5 (38%) 85%/15% 20 (8–52) 27 (17–58)

12 (92%) 0 1/13 (8%)

Cutaneous vasculitides

Literature search This series

ANCA-associated vasculitides

Literature search This series

144 70 (49%) 74 (51%) 82%/18% 21 (10–56)

5 66 4 (80%) 43 (65%) 1 (20%) 23 (35%) 40%/60% 52%/48% 38 (26–59) 25 (2–80)

8 2 (25%) 6 (75%) 88%/12% 31 (5–58)

23 (12–52)

49 (21–59)

52 (17–71)

87/127 (69%) 22/127 (17%) 18/127 (14%)

3 (60%) 2/5 (40%) 0

26 (2–80)

32/55 (58%)b 14/55 (25%) 9/55 (16%)

8 (100%) 0 0

Other vasculitides

Literature search This series

Literature search

19 12 (63%) 7 (37%) 58%/42% 29 (10–53)

6 4 (67%) 2 (33%) 50%/50% 21.5 (13–48)

78 45 (58%) 33 (42%) 44%/56% 22 (10–59)

35 (16–78)

31 (10–49)

31 (10–59)

17 (89%) 2 (11%) 0

3 (50%) 3 (50%) 0

45/59 (76%) 10/59 (17%) 4/59 (7%)

ANCA, anti-neutrophil cytoplasmic antibody; IBD, inflammatory bowel disease. a The temporal relationship between the diagnoses of IBD versus vasculitis was reported or available for 126 cases of large-vessel vasculitides, 55 cases of cutaneous vasculitides, and 59 cases of other vasculitides. b Includes cases of possible drug-induced vasculitis.

A. Sy et al. / Seminars in Arthritis and Rheumatism 45 (2016) 475–482

481

Table 3 Characteristics of patients with Takayasu arteritis (TAK) and an associated IBD from this series and the literature, and with TAK but no IBD from the Vasculitis Clinical Research Consortium (VCRC) longitudinal study Characteristic

Female/male, no. Age at diagnosis of TAK, years, median (range) Age r50 years old, no. (%) Ethnicity, no. (%) Caucasian Asian African American Middle Eastern Native Indian

Patients from this series with TAK and IBDa, N ¼ 12 11/1 19 (8–52) 11 (92) 8 (67) 3 (25) 1 (8) 0 0

Clinical manifestations at time of TAK diagnosis, no. (%) Claudication of extremities Decreased brachial pulse Blood pressure difference between arms Bruit over subclavian artery or aorta Renal hypertension Fever/weight loss/fatigue Abdominal pain, diarrhea Angina/dyspnea Carotidynia Headache/dizziness Arthralgia

8 9 7 5 4 7 4/5 2 1 4 2

Duration of follow-up, years, median (range)

4.7 (0.75–28)

Outcomes, no. with available data, no. (%) Remission of vasculitis at last follow-up Stroke Myocardial infarction Deaths

(67) (75) (58) (42) (33) (58) (80) (17) (8) (33) (17)

n ¼ 8 5 (63) 0 0 0

Patients from the literature with TAK and IBD, N ¼ 133

Patients from this series with TAK and no IBD (VCRC), N ¼ 152

100/23b 23 (10–69) 104/105 (99)b

143/9 31 (4–63)c 139 (91)c

n ¼ 62b 11 (18) 45 (73) 3 (5) 3 (5) 0

127 (84)c 14 (9)c 7 (5%) 0c 4 (3)

n ¼ 67b 19 (28) 36 (54) 35 (52) 35 (52) 9 (13) 31 (46) 25 (37) 14 (21) 14 (21) 11 (16) 9 (13) 0.4 (0.1–28) b

n ¼ 68 55 4 1 6

(81) (6) (1) (9)

96 92 70 77 20 42 6 14 31 12 37

(63)c (61) (46) (51) (13) (28)c (4)c (9)c (20) (8)c (24)

6.25 (0.25–31)c n ¼ 122 108 (89) 9 (7) 4 (3) 0c

ACR, American College of Rheumatology; BP, blood pressure; IBD, inflammatory bowel disease; TAK, Takayasu arteritis; VCRC, Vasculitis Clinical Research Consortium. a Patients from this series encompass patients from 3 different sources: the VCRC Longitudinal Studies cohorts, the Canadian Vasculitis research network centers and the Mount Sinai Hospital–University of Toronto referral center for IBD. b Complete clinical and demographics information was not reported for all cases from the literature. c p o 0.05 comparing patients with TAK and IBD from this series and the literature versus those with TAK but no IBD from the VCRC.

other, especially when the gastrointestinal manifestations are isolated [3,6,12]. The comparison of the characteristics of the patients with TAK with or without IBD also adds to the available literature. This study has some limitations to consider. It is retrospective, and hence certain detailed clinical information such as the initial clinical features that led to a suspicion of vasculitis in a patient with underlying IBD is not always available from the VCRC database. However, a standardized recording form was used and physicians were contacted as needed to obtain the missing information. While the literature review was comprehensive, some cases of patients with IBD and vasculitis that were embedded within general case series of patients with vasculitis may have been missed. However, these cases would have likely lacked detailed clinical information relevant to the study and, therefore, would have had minimal impact on the findings. In addition, the diagnosis of IBD and vasculitis often relied on authors’ statements, rather than strictly using classification criteria [4,5]. Finally, in the absence of a longitudinal multicentric IBD cohort, the exact frequency of vasculitis in patients with IBD could not be determined. This case series and literature review highlight the likely noncoincidental association of IBD and vasculitis, although the association remains rare. TAK is the most common, but not exclusive type, of vasculitis observed with IBD. The cooccurrence of these diagnoses most often arises in young patients with already established IBD and at a time when the IBD is usually not active. Clinicians treating IBD should have a high index of suspicion for vasculitis in patients with unusual

symptoms and should have a low threshold to screen patients for the possibility of vasculitis. The treatment and outcomes of patients with vasculitis and IBD do not overtly differ from those of patients with “primary” vasculitis and most of the treatments used for vasculitis are also effective against IBD. Future research may help to clarify the complex relationships and pathogenic mechanisms shared by these 2 diseases. Appendix A. Supplementary information Supplementary material cited in this article is available online at http://dx.doi.org/10.1016/j.semarthrit.2015.07.006.

References [1] Jennette JC, Falk RJ, Bacon PA, Basu N, Cid MC, Ferrario F, et al. 2012 Revised International Chapel Hill Consensus Conference Nomenclature of Vasculitides. Arthritis Rheum 2013;65:1–11. [2] Camilleri M, Pusey CD, Chadwick VS, Rees AJ. Gastrointestinal manifestations of systemic vasculitis. QJM 1983;52:141–9. [3] Pagnoux C, Mahr A, Cohen P, Guillevin L. Presentation and outcome of gastrointestinal involvement in systemic necrotizing vasculitides: analysis of 62 patients with polyarteritis nodosa, microscopic polyangiitis, Wegener granulomatosis, Churg–Strauss syndrome, or rheumatoid arthritis-associated vasculitis. Medicine 2005;84:115–28. [4] Conrad K, Roggenbuck D, Laass MW. Diagnosis and classification of ulcerative colitis. Autoimmun Rev 2014;13:463–6. [5] Laass MW, Roggenbuck D, Conrad K. Diagnosis and classification of Crohn’s disease. Autoimmun Rev 2014;13:467–71. [6] Reny JL, Paul JF, Lefebvre C, Champion K, Emmerich J, Bletry O, et al. Association of Takayasu’s arteritis and Crohn’s disease. Results of a study on

482

[7]

[8]

[9]

[10]

[11]

[12]

[13] [14]

[15]

[16] [17] [18]

A. Sy et al. / Seminars in Arthritis and Rheumatism 45 (2016) 475–482

44 Takayasu patients and review of the literature. Ann Med Interne 2003;154:85–90. Judah J, Hammond C, Polyak S, Drane W, Valentine J. The coexistence of Crohn’s disease and Takayasu arteritis: diagnosis and treatment of combined disease in three patients. Pract Gastroenterol 2009;3:50–8. Todini AR, Heinzmann MM, Antignani PL, Paiella ML. Association between Takayasu’s arteritis and Crohn’s disease in two young women: case reports. J Mal Vasc 1999;24:373–6. Vaszar LT, Orzechowski NM, Specks U, Ytterberg SR, Loftus EVEV Jr, Mark EJ, et al. Coexistent pulmonary granulomatosis with polyangiitis (Wegener granulomatosis) and Crohn disease. Am J Surg Pathol 2014;38:354–9. Burns AM, Walsh N, Green PJ. Granulomatous vasculitis in Crohn’s disease: a clinicopathologic correlate of two unusual cases. J Cutan Pathol 2010;37:1077–83. Unnikrishnan A, Azodi S, Ansari N, Brown M, Kamnetz J, Uchiyama RC. PR3ANCA related cerebral vasculitis in ulcerative colitis presenting with orbital involvement: a case report with review of literature. Case Rep Rheumatol 2014;2014:582094. Kasuga A, Mandai Y, Katsuno T, Sato T, Yamaguchi T, Yokosuka O. Pulmonary complications resembling Wegener’s granulomatosis in ulcerative colitis with elevated proteinase-3 anti-neutrophil cytoplasmic antibody. Intern Med 2008;47:1211–4. Grigg EL, Kane S, Katz S. Mimicry and deception in inflammatory bowel disease and intestinal behcet disease. Gastroenterol Hepatol 2012;8:103–12. Grayson PC, Cuthbertson D, Carette S, Hoffman GS, Khalidi NA, Koening CL, et al. New features of disease after diagnosis in 6 forms of systemic vasculitis. J Rheumatol 2013;40:1905–12. Fries JF, Hunder GG, Bloch DA, Michel BA, Arend WP, Calabrese LH, et al. The American College of Rheumatology 1990 criteria for the classification of vasculitis. Summary. Arthritis Rheum 1990;33:1135–6. Osman M, Aaron S, Noga M, Yacyshyn E. Takayasu’s arteritis progression on anti-TNF biologics: a case series. Clin Rheumatol 2011;30:703–6. Hall S, Barr W, Lie JT, Stanson AW, Kazmier FJ, Hunder GG. Takayasu arteritis. A study of 32 North American patients. Medicine 1985;64:89–99. Minami N, Nakase H, Yoshino T, Yamada S, Toyanaga T, Honzawa Y. Effect of infliximab on inflammatory bowel disease with Takayasu

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26] [27]

[28]

[29]

arteritis: case series and review of the literature. Clin J Gastroenterol 2013;6:226–30. Takahashi N, Tanabe K, Sugamori T, Sato M, Kitamura J, Sato H, et al. Association between Takayasu arteritis and ulcerative colitis—case report and review of serological HLA analysis. Med Sci Monit 2011;17:CS81–4. Terao C, Matsumura T, Yoshifuji H, Kirino Y, Maejima Y, Nakaoka Y, et al. Takayasu arteritis and ulcerative colitis—high concurrence ratio and genetic overlap. Arthritis Rheumatol 2015;67(8):2226–32. Wakefield AJ, Sankey EA, Dhillon AP, Sawyerr AM, More L, Sim R, et al. Granulomatous vasculitis in Crohn’s disease. Gastroenterology 1991; 100(5 Pt 1):1279–87. Humbert S, Guilpain P, Puechal X, Terrier B, Riviere S, Mahr A, et al. Inflammatory bowel diseases in anti-neutrophil cytoplasmic antibodyassociated vasculitides: 11 retrospective cases from the French Vasculitis Study Group. Rheumatology (Oxford) 2015 http://www.ncbi.nlm.nih.gov/ pubmed/26106214 [Epub ahead of print]. Zikou AK, Kosmidou M, Astrakas LG, Tzarouchi LC, Tsianos E, Argyropoulou MI. Brain involvement in patients with inflammatory bowel disease: a voxelbased morphometry and diffusion tensor imaging study. Eur Radiol 2014;24:2499–506. Katsanos AH, Kosmidou M, Giannopoulos S, Katsanos KH, Tsivgoulis G, Kyritsis AP, et al. Cerebral arterial infarction in inflammatory bowel diseases. Eur J Intern Med 2014;25:37–44. Horai Y, Satoru O, Lapalme-Remis S, Sumiyoshi R, Nakashima Y, Suzuki T, et al. Takayasu arteritis developing during treatment of ulcerative colitis with infliximab. Mod Rheumatol 2013;23:572–6. Rustagi T, Majumder S. Crohn’s-Takayasu’s arteritis overlap with hypercoagulability: an optimal milieu for ischemic stroke. J Dig Dis 2011;12:142–6. Katoh N, Kubota M, Shimojima Y, Ishii W, Matsuda M, Akamatsu T, et al. Takayasu’s arteritis in a patient with Crohn’s disease: an unexpected association during infliximab therapy. Intern Med 2010;49:179–82. Ramos-Casals M, Brito-Zeron P, Munoz S, Soria N, Galiana D, Bertolaccini L, et al. Autoimmune diseases induced by TNF-targeted therapies: analysis of 233 cases. Medicine 2007;86:242–51. Sokumbi O, Wetter DA, Makol A, Warrington KJ. Vasculitis associated with tumor necrosis factor-alpha inhibitors. Mayo Clin Proc 2012;87:739–45.